Needle-induced cavitation: A method to probe the local mechanics of brain tissue

Abstract

Traditional mechanical characterization of extremely soft tissues is challenging given difficulty extracting tissue, satisfying geometric requirements, keeping tissues hydrated, and securing the tissue in an apparatus without slippage. The heterogeneous nature and structural complexity of brain tissues on small length scales makes it especially difficult to characterize. Needle-induced cavitation (NIC) is a technique that overcomes these issues and can mechanically characterize brain tissues at precise, micrometer-scale locations. This small-scale capability is crucial in order to spatially characterize diseased tissue states like fibrosis or cancer. NIC consists of inserting a needle into a tissue and pressurizing a fluid until a deformation occurs at the tip of the needle at a critical pressure. NIC is a convenient, affordable technique to measure mechanical properties, such as modulus and fracture energy, and to assess the performance of soft materials. Experimental parameters such as needle size and fluid flowrate are tunable, so that the end-user can control the length and time scales, making it uniquely capable of measuring local mechanical properties across a wide range of strain rates. The portable nature of NIC and capability to conduct in vivo experiments makes it a particularly appealing characterization technique compared to traditional methods. Despite significant developments in the technique over the last decade, wide implementation in the biological field is still limited. Here, we address the limitations of the NIC technique specifically when working with soft tissues and provide readers with expected results for brain tissue. Our goal is to assist others in conducting reliable and reproducible mechanical characterization of soft biomaterials and tissues.

Introduction

Needle-induced cavitation (NIC) was developed to measure local mechanical properties in soft, heterogeneous materials, and has been widely used^1–12. NIC is a cavitation rheology technique¹³ used to characterize the local elastic and fracture properties of biological tissues and soft networks. NIC can quantify the elastic modulus $E$ and fracture energy $G_c$ of a material or tissue¹⁴, and compare the relative mechanical properties across samples, across locations, and/or over time. NIC is performed at the tip of an embedded needle, thus enabling the measurement of properties within highly localized volumes (Fig. 1a). The length scale of the probed volume is defined by the outer radius of the needle tip, making it an exciting technique for characterizing biological tissues containing structures on multiple size scales (μm to cm). This is particularly useful for precisely mapping the modulus of a heterogeneous tissue spatially¹(Fig 1b). Compared to techniques capable of measuring moduli on the same length scale, NIC is quick, uses minimal equipment, and requires no specific tissue preparation. The capability to measure mechanical properties in situ with spatial resolution provides insight into diseased states such as fibrotic lung diseases,^15,16 brain altering disorders like epilepsy and cerebral palsy^17,18, or cancer tumor progression^19,20, which will contribute to the development of better in vitro biomaterials to recapitulate the mechanical states during disease.

Figure 1. Workflow for mouse brain NIC.

a Schematic of a portable NIC setup with all tubing, syringe, pressure sensor, and needle. b Example of mapping regional NIC mechanical characterization of mouse brain using the Allen Mouse Brain Atlas. c Needle insertion with representative force curve to ensure puncture of needle into brain tissue. d Representative pressure curve for NIC of mouse brain with critical pressure $P_c$ (red circle) used to calculate the Young’s elastic modulus $E$ given an interfacial tension between brain and NIC fluid $\gamma$ and needle outer radius $r$. Made in BioRender.

NIC has been used to characterize intact and in situ soft biological tissues such as: eye lens⁵, vitreous⁶, cell spheroids²¹, bone marrow²², skin²³, lung²⁴, and brain^25,26. Given that NIC is relatively new to the field, researchers that hope to implement this technique may face a significant barrier when attempting to identify best practices.

NIC minimally consists of a needle, pressure sensor, and a method to control the flow rate of a fluid through the needle into a tissue. The pressure vs. time curves obtained during pressurization at the needle tip can be used to obtain the modulus of a tissue (Fig 1d). Cavitation is typically modeled as the elastic expansion of a spherical void within an incompressible solid. Assuming the solid is Neo-Hookean gives

$$P-P_{surf}=\frac{5}{6}E-\frac{2E}{3\lambda }-\frac{E}{6{\lambda }^4}$$ (1)

where $P$ is pressure, $P_{surf}$ is the Laplace pressure, $E$ is Young’s elastic modulus, and $\lambda$ is stretch ratio defined as the current radius divided by the initial radius^27,28. When deformation is driven by elasticity, a simplified version of Eq. 1 relates the critical pressure $P_c$ when the expansion of the void suddenly grows, and $\lambda \to \infty$, to the elastic modulus $E$ of the sample

$$P_c=\frac{2\gamma }{r}+\frac{5}{6}E$$ (2)

where $r$ is the radius of the spherical void. For NIC, $r$ is set by the radius of the needle. Therefore, the peak pressure, which is typically equal to $P_c$, should have a linear dependence on the inverse needle radius $1\text{/}r$, and the materials property $E$ is related to the critical pressure $P_c$.

A typical, NIC setup is shown in Fig. 2a, with components incorporated to monitor the force and displacement of the needle, measure the pressure of the fluid within the needle, and visualize deformations at the needle tip. Although only a needle, fluid (gas or liquid), and pressure sensing equipment are necessary for performing NIC, including all the elements in Fig. 2a enables access to additional data that can aid the establishment of best practices for NIC and improve data analysis^{3,8,11,12,29–32}. Aside from a benchtop NIC setup, a portable NIC setup may be ideal for transport and includes only a needle, pressure sensor, and syringe pump (Fig. 2b).

Figure 2. NIC experimental setup.

Schematic a of a needle-induced cavitation setup capable of monitoring pressure at the needle tip, measuring needle force and displacement, and visualizing the needle tip during testing. Photos of a b portable needle-induced cavitation setup. *shows useful but not necessary equipment.

Here, we provide detailed information and a case study to aid the use of NIC to quantify the localized moduli of soft tissues, which can be difficult to measure with traditional techniques. We use a portable NIC setup, with the goal of addressing the nuances and limitations associated with NIC and assisting others in conducting straightforward, reproducible mechanical NIC characterization experiments on soft biological tissues. The following experimental guidelines will aid NIC users in choosing parameters that will work well for characterizing various biological or soft materials, and we address potential complications with suggested confirmation steps.

Results

Needle Choice and Flowrate are Brain Tissue Specific.

Choosing a needle size for NIC requires consideration of the length scale of measurement desired, and needles can vary from 1 μm to greater than 1.5 mm (Fig. 3a) in outer diameter. Based on the analysis of Pavlovsky et al., the maximum needle outer radius $R_{Max}$ for a given sample size can be estimated $\left(R_{Max}=\text{h/}20\right)$¹⁰ using the height of the sample $h$, assuming a cylindrical shaped sample given a sample volume $V_{Sample}$ (Fig. 3b). Mouse brain has total volume of ~400 mm³, so the maximum needle outer radius, ${\text{R}}_{\text{Max}}$, for NIC experiments is 0.25 mm (Fig 3b)³³. Therefore, we chose to proceed with a 27 G (413 μm OD) needle. As a point of reference for smaller volume samples, NIC has been performed on cell spheroids that are ~400 μm in diameter²¹ with glass needles that are 5 – 30 μm in diameter.

Figure 3. Choosing needle size/geometry and flowrate for mouse brain NIC.

a Multiple needle tip styles and sizes are available for NIC. Scale bar 5 mm. Schematic of needle retraction with a flat or beveled needle. b Plot showing the maximum outer needle radius ${\text{R}}_{\mathrm{Max}}$ needed to test samples of different volumes. Orange dot indicates needle size restrictions for NIC on mouse brain. c Plot of critical pressure versus retraction distance for mouse brain using a flat or beveled needle. d Effect of NIC flowrate on critical pressure for mouse brain.

NIC was developed using flat needles¹; however, when using a flat needle it is necessary to retract the needle after insertion to reduce the residual compressive strain below the needle tip³. Residual strain impacts the pressurization process and can increase the observed critical pressure by an order of magnitude³. For a flat needle, the required retraction distance (Fig. 3a) varies by sample, for example, >0.5 mm for mouse brain (Fig. 3c). The critical retraction distance depends on the amount of compressed material below the tip, which is connected to the elasticity and fracture energy of the material as well as the adhesion energy between the needle and sample³. Beveled needles concentrate stress at the needle tip reducing the amount of compressed material below the needle, and puncture occurs in tissues at lower displacements as compared to flat needles. This makes beveled needles an ideal choice for NIC on mouse brain, however flat needles are acceptable if the retraction distance is sufficient to mitigate residual strains that may develop during puncture (Fig. 3c).

Fluid flowrate controls expansion at the needle tip, thus controlling strain rate. Fluid flowrate can be determined by an estimated strain rate of interest or by a desired time to complete each experiment (e.g. within 1 minute). One consideration for soft tissues is strain rate dependency or viscoelasticity^34,35. Brain tissue shows a strain rate dependency at low flowrates (25–50 μL/min). However, the strain rate effects level off around 250 μL/min, making that a suitable flowrate for NIC on mouse brain (Fig. 3d). Having control over the fluid flowrate makes this experimental setup convenient to probe mechanics at varying strain rates, however the leveling off of critical pressures at higher flowrates is unique to NIC and not observed with other mechanical testing techniques. This may indicate limits in the NIC strain rate regime for brain tissue at short length scales.

Water is Ideal NIC Fluid for Brain Tissue

Considering the pressure-growth relationship for a spherical void in an incompressible Neo-Hookean material, the critical pressure $P_c$ at which cavitation occurs is related to the material properties according to Eq. 2. The interfacial tension $\gamma$ between the fluid and sample will increase the observed critical pressure $P_c$ ¹. Therefore, choosing an appropriate injection fluid requires consideration of the interfacial tension between the fluid and sample, as well as the viscosity of the injected fluid.

To minimize the effects of interfacial tension on the critical pressure for biological tissues, NIC can be performed with water, which is highly miscible with brain tissue, given brain consists of 80% water. By choosing a fluid with negligible interfacial tension with the material being tested, the critical pressure is directly related to the elastic modulus. For mouse brain NIC experiments conducted with water (Fig. 4b), equation 2 simplifies to:

$$P_c=\frac{5}{6}E$$ (3)

Figure 4. Water is appropriate NIC fluid for brain tissue.

a Plot of elastic modulus versus the time at which the critical cavitation event occurs ${\text{t}}_{\text{c}}$. The solid lines mark out the boundary above which a fluid has an appropriate viscosity as a cavitating agent. The star shows that a test run for 30 s in a sample with $E=1000$ Pa water would be a suitable fluid while glycerol would not. b Modulus is calculated from the critical pressure (red circles) of multiple regions of mouse brain for water NIC.

To create a fully incompressible NIC system, the pressure sensor, tubing, and fittings can be filled with water before connecting (Fig. S2). An initial setup of the syringe, pressure sensor, tubing, and needle for NIC will typically take longer than the ensuing experiments.

Selecting a fluid with an appropriate viscosity depends on balancing the resistance due to material response with that from viscous drag during injection. In the supplementary material, a simple balance is developed in order to estimate the viscosity at which the drag forces will be equivalent to those from the material response. The results of this analysis are shown in Figure 4a where $E$ is plotted against the time at which the critical event $t_c$ occurs. The solid lines mark a boundary below which the viscosity of a given fluid makes it a poor choice for NIC. For example, if a typical test takes 30 s on a sample where $E=1000$ Pa, as shown by the star in Figure 4a, water would be an appropriate fluid while glycerol would not be an appropriate fluid.

In NIC experiments, the elastic modulus $E$ was estimated by Eq. 2 in four different regions of the brain: 3.6 ± 1.1, 5.3 ± 1.4, 6.4 ± 1.0, and 3.3 ± 1.0 kPa for the cortex, thalamus, hypothalamus, and cerebellum, respectively (n = 10 – 21) (Fig. 4b). These values are similar to the cortex values measured using volume-controlled needle-induced cavitation rheology²⁵. Using high sensitivity pressure sensors on soft materials results in noisy data, however modulus calculations are only dependent on the maximum critical pressure, so no data smoothing or filtering is necessary.

Needle Insertion Monitoring to Ensure Puncture

The needle insertion process involves inserting a needle into a sample to a specific depth, which can be controlled manually or with actuators. Prior to NIC, needle insertion is mapped using the Allen Mouse Brain Atlas to probe specific regions of the brain (Fig. 5a-b). It is helpful to create or use grids for large tissues when preparing biological samples for NIC to ensure precise regionals mechanical characterization. Although NIC has the capability to measure mechanical properties of intact soft tissues within or outside the body, it is important to consider how the properties of a tissue change from in situ, ex vivo, and in vivo. Ex vivo biological samples are prepared by excising the tissue and immersing it in an ice cold, physiological solution, such as Hank’s balanced salt solution (HBSS) for brain tissue to minimize effects due to drying and to preserve the integrity of the sample.

Figure 5. Preparing biological samples for NIC.

Allen Mouse Brain Atlas mapping for NIC needle insertion in a coronal and b sagittal view. c Custom 3D printed holder to secure mouse brain during NIC experiments. Scale bar 5 cm. d Representative force monitoring to ensure needle puncture into brain sample. e Representative pressure curve for NIC on mouse brain.

By using a container that is very close in size to the tissue sample being tested, the tissue remains in one place allowing better control over needle insertion. Designing and 3D printing a custom container for specific tissue is a straightforward and inexpensive way to prevent any slippage during NIC (Fig. 5c). An example file to hold small mouse tissues is included in the supplemental material. For in situ or in vivo experiments, it is important to secure the samples to avoid slippage, potentially guiding needle insertion using microscopy or surgical loupes.

While inserting the needle, it is critical to choose a depth that will ensure the bevel is below the surface of the tissue. If the needle is not fully inserted into the sample, fluid leakage can occur resulting in an inaccurately low critical pressure. Brain tissue provides little resistance to needle insertion, making visual confirmation of needle insertion difficult. Therefore, a force transducer is helpful to visualize when needle puncture occurs, as is indicated by a large decrease in force (Fig. 5d)^36–40.

Puncture can be more challenging to confirm when force transducers are unable to be integrated, for example in a portable cavitation device. Here, we often use visual observation of the recoil of the surface of the tissue or gel to indicate puncture, and we can also rely upon the guidance provided by the previous work on puncture. For example, Fakhouri et al found that the critical depth of puncture was approximately 13 times the radius of curvature for the penetrating needle³⁶; and Rattan et al found that the critical depth of penetration was approximately 27 times the needle radius of curvature⁴¹. Another confirmation that the needle has punctured the tissue is a rise in pressure once the syringe pump has been turned on post needle insertion (Fig. 5e).

NIC fluid is pressurized until a cavitation event occurs at the tip of the needle, associated with an observed pressure change: a drop off or leveling off of pressure after the initial increase¹ (Fig. 5e). This expansion can be driven by either a cavitation mechanism or a fracture mechanism. Distinguishing cavitation from fracture in biological tissues can be complicated. However, NIC can be used to indicate relative changes in while NIC can measure elastic modulus $E$ without differentiating cavitation from fracture by comparing the critical pressures.

Discussion

Many references have compared the elastic properties measured via traditional techniques such as shear rheometry^{1,2,7,8,22,42–44}, uniaxial extension⁹, or indentation^{3,4,22,24,45,46} to those characterized by NIC and found similar results. We observe that NIC and indentation on mouse brain cortex result in similar elastic moduli (Fig. S4). Most alternative mechanical characterization techniques (e.g., uniaxial tension/compression and shear rheology) are excellent at reporting bulk moduli of tissue^47–59, while NIC can measure an elastic modulus at localized regions of a tissue without dissection, for example the cortex or thalamus region of brain tissue. The length scale of measurement is determined by the needle size and retraction distance (Fig. 3a). In fact, performing NIC with multiple needle sizes on a heterogeneous material would allow for measurements across varying length scales. NIC has been used to characterize homogeneous materials such as polyvinyl alcohol, polyethylene oxide, and poly-(methyl methacrylate) based triblock polymers using needle sizes ranging from single microns to 100’s of microns^60–62. There has been recent focus on developing characterization techniques to measure local mechanical properties of tissues including indentation^63–66 and atomic force microscopy (AFM)^67–69, micropipette aspiration⁷⁰, and magnetic resonance elastography^71,72, but these techniques typically require expensive equipment and extensive training. NIC is extremely accessible due to the straightforward procedure and simple experimental setup. NIC allows for multiple tests within the same tissue sample to directly compare different areas, or even fibrotic tissue vs. neighboring healthy tissue.

One significant benefit of NIC is that it can be performed on in situ (e.g. bone marrow in bone or vitreous in the eye^6,22), ex vivo (brain and lung ^24,26), and in vivo (skin²³) tissues at multiple length scales ranging from bulk to subcellular. NIC has revealed that the elastic modulus of tissues change when removed from their native environments^6,22 and tissue moduli have clear spatial heterogeneity, also captured using the small needle sizes with NIC. Additionally, many tissues are viscoelastic, and NIC has recently been used to probe viscoelastic properties^30,73,74 at low strain rates (10⁻⁴ - 10⁻¹ Hz)¹³. For materials with a Young’s elastic modulus above 100 kPa, such as bone, NIC would not be an appropriate characterization technique. Although different needle sizes allow for characterization across multiple length scales, NIC is limited in length scale based on the chosen needle size, making measurements across a large tissue potentially time consuming. Finally, NIC is a powerful tool in characterizing biomaterials, especially for small volume systems, and is compatible with cell encapsulated matrices, both because it is portable and can reside within a biosafety cabinet. This has implications for researchers developing mechanically functional materials⁷⁵, and characterizing healthy and disease-specific^76–78 tissue-mimicking materials. NIC has been embraced by the gel mechanics and biomechanics communities in characterizing soft materials ranging from 100 Pa to 100 kPa at low strain rates (10⁻⁴ – 10³ s⁻¹)¹³.

One concern with using NIC is determining whether or not the expansion process observed should be modeled as a cavitation or fracture process, which is complicated by the fact that expansion initiated by cavitation can transition into fracture driven expansion^8,13,79. This is complicated in tissue because observing cavity morphology after expansion is impossible in opaque tissues. The most rigorous means of experimentally determining whether cavitation has occurred is to plot $P_c$ against $R_0^{-1}$ to observe the scaling of critical pressure for multiple needle sizes²⁹. If this plot scales linearly or as a square root then a cavitation or fracture process is driving the initial expansion, respectively⁴⁵. Distinguishing between cavitation and fracture is important because the critical cavitation pressure represents the upper threshold pressure value for the onset of expansion. This means that a fracture-initiated process modeled as a cavitation process will underestimate the value of $E$ in NIC. Although NIC can be used to quantitatively determine $E$ and $G_c$ in cases where cavitation can be distinguished from fracture, it can also be used to indicate relative changes in $E$ without differentiating cavitation from fracture. In cases where cavitation occurs, this follows from Eq. 1 which shows that $P_c$ scales directly with $E$. When fracture occurs, $P_{Fracture}$ also has a robust positive scaling with $E$ so it is not always vital to distinguish whether the critical event is driven by cavitation or fracture to comment on the trend in modulus. Put simply, if the experimentally observed critical pressure $P_c$ increases between samples one can confidently say that $E$ increases between the samples. This makes NIC an attractive option when attempting to make relative material comparisons without distinguishing between cavitation and fracture.

When considering NIC usage for in vivo applications, it will be important to better understand the extent of damage from needle insertion and the cavitation process itself. Given that needle puncture occurs in many biomedical instances, from vaccinations or epidurals, the damage from needle insertion should be considered minimal, however the extent of damage from cavitation is an area of research that is underexplored. NIC has the potential for clinical applications, such as measuring tissue/tissue interface strength, and has already been proposed as a nondestructive method of assessing skin mechanics during wound healing and disease progression²³.

Conclusion

NIC is a powerful characterization tool with the potential to characterize intact in situ, ex vivo, and in vivo soft biological tissues. Because of the portable nature, minimal sample preparation, and time and length scale tunability, NIC is an attractive method to characterize materials compared to traditional methods. We provide a case study for NIC on mouse brain and discuss technique optimization for a variety of applications in an effort to make NIC accessible to researchers working with soft biomaterials and tissues. Through our experience developing and optimizing NIC for soft biological tissues, we provide guidelines for choosing needle size, retraction, fluid flowrate, and fluid type, with suggestions on preparing samples and analyzing data with the goal that others can implement this system in their own labs. In addressing the nuances and limitations of NIC on biological tissues, we expect that others can conduct straightforward, reproducible mechanical characterization experiments.

Materials and Methods

Animals.

Male and female BALB/c, a mix of homozygous (−/−) or heterozygous (+/−) nude mice (6–8 weeks old; The Jackson Laboratory, Bar Harbor, ME USA) were used. The mice were housed in a pathogen-free environment with a 12:12-hour light:dark cycle and controlled humidity and temperature, with unlimited access to food and water. We first sacrificed mice, excised their brains 4–10 at a time, and stored the tissues in ice cold 1X Hank’s balanced salt solution (Gibco, Waltham, MA USA) prior to NIC. All protocols applied in the experiments were approved by the Institute of Animal Care and Utilization Committee at University of Massachusetts Amherst.

Needle-Induced Cavitation (NIC).

The collected mouse brains were stored in ice cold 1X Hank’s balanced salt solution (Gibco) until NIC, and all mouse brain experiments were conducted within 30 minutes post-harvesting. For the NIC experiment, the cavitation fluid was distilled. Just prior to testing, brain tissue was placed in a custom 3D printed (uPrintSE+, AET Labs, Essex, MA USA) specimen holder. Control of the pressure was achieved using a 3 mL syringe (11.25 mm ID) in an WPI syringe pump (World Precision Instruments). During NIC experiments, the real time pressure was monitored using a Px409–015 GUSBH (Omega, Norwalk, CT USA) and interfaced with a custom LabView program. The depth of needle insertion was controlled using an actuator (Texture Technologies, Hamilton, MA USA). Beveled 27 G metal needles with {$R_{in}$, $R_{out}$} = {210, 413} μm were used with no retraction. The interfacial tension was assumed to be zero. The pressurization rate was 250 μL/min.

Abbreviations

NIC

Needle-induced cavitation

PTFE

polytetrafluoroethylene

AFM

atomic force microscopy